EUVIMEDEUVIMED entdecken
Crossref

PRICE: A Pretrained Model for Cross-Database Cardinality Estimation

Tianjing Zeng, Junwei Lan, Jiahong Ma, Wenqing Wei, Rong Zhu, Yingli Zhou, Pengfei Li, Bolin Ding, Defu Lian, Zhewei Wei, Jingren Zhou

Proceedings of the VLDB Endowment · 2024-11 · 18 (3) · 637-650

DOI: 10.14778/3712221.3712231 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the VLDB Endowment, online erschienen am 07.04.2025. Verfasst von Tianjing Zeng und 10 weiteren Autor:innen. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 87 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cardinality estimation (CardEst) is essential for optimizing query execution plans. Recent ML-based CardEst methods achieve high accuracy but face deployment challenges due to high preparation costs and lack of transferability across databases. In this paper, we propose PRICE, a PRetrained multI-table CardEst model, which addresses these limitations. PRICE takes low-level but transferable features w.r.t. data distributions and query information and elegantly applies self-attention models to learn meta-knowledge to compute cardinality in any database. It is generally and adaptively applicable to any unseen new database to attain high estimation accuracy, while its preparation cost is as little as the basic one-dimensional histogram-based CardEst methods. Moreover, PRICE can be finetuned to further enhance its performance on any specific database. We pretrained PRICE using 30 diverse datasets, completing the process in about 5 hours with a resulting model size of only about 40MB. Evaluations show that PRICE consistently outperforms existing methods, achieving the highest estimation accuracy on several unseen databases and generating faster execution plans with lower overhead. After finetuning with a small volume of database-specific queries, PRICE could even find plans that were very close to the optimal ones. Meanwhile, PRICE is generally applicable to different settings such as data updates, data scaling, and query workload shifts.

Zitieren

APA

Zeng, T., Lan, J., Ma, J., Wei, W., Zhu, R., Zhou, Y., Li, P., Ding, B., Lian, D., Wei, Z., & Zhou, J. (2024). PRICE: A Pretrained Model for Cross-Database Cardinality Estimation. Proceedings of the VLDB Endowment, 18(3), 637-650. https://doi.org/10.14778/3712221.3712231

Vancouver

Zeng T, Lan J, Ma J, Wei W, Zhu R, Zhou Y, et al. PRICE: A Pretrained Model for Cross-Database Cardinality Estimation. Proceedings of the VLDB Endowment. 2024;18(3):637-650. doi:10.14778/3712221.3712231

BibTeX
@article{Zeng2024,
  author = {Zeng, Tianjing and Lan, Junwei and Ma, Jiahong and Wei, Wenqing and Zhu, Rong and Zhou, Yingli and Li, Pengfei and Ding, Bolin and Lian, Defu and Wei, Zhewei and Zhou, Jingren},
  title = {PRICE: A Pretrained Model for Cross-Database Cardinality Estimation},
  journal = {Proceedings of the VLDB Endowment},
  year = {2024},
  volume = {18},
  number = {3},
  pages = {637--650},
  doi = {10.14778/3712221.3712231}
}
Kostenlos starten →